L’accident mortel en Tesla aurait peut-être pu être évité

L’accident mortel en TESLA aurait peut-être pu être évité grâce à des modules logiciels de NEXYAD

English translation

par NEXYAD, août 2016

Les passions se déchaînent autour de la question des véhicules autonomes, ou semi-autonomes. Récemment, une personne a perdu la vie dans son véhicule TESLA alors qu’il était en mode auto-pilote.
NEXYAD a étudié la sécurité routière pendant une vingtaine d’année, et nous donnons quelques éléments de réflexion sur ce type d’accident.

Les chaînes de traitement de l’information des systèmes auto-pilote, driverless, etc… allant de la perception de l’environnement jusqu’à la prise de décision et à la gestion automatique des actionneurs, sont généralement très bien conçues, et mettent en œuvre des modules performants. Mais cela ne suffit pas à rendre nul le risque d’accident. En effet, pour traiter ce risque, il manque une chaîne parallèle (et indépendante) de « monitoring ».
C’est pour bien comprendre cette nécessité, il faut tout d’abord appréhender le niveau de complexité d’une scène routière vue par une caméra.

Le livre blanc à lire ici

Accident fatalities in a TESLA car

Accident fatalities in a TESLA car might have been avoided by using software modules of NEXYAD :
the time for monitoring circuit has come.

Version française

By NEXYAD

Processing circuit, informing auto-pilot systems, control, etc … from perception, data fusion, decision-making, and automatic control of actuators, are usually very well designed, and based on high-performance modules. But unfortunately, this is not enough to void the risk of accidents. Indeed, for the treatment of this risk, it lacks a parallel circuit (parallel and independent) called “monitoring” circuit.
To understand this need for a monitoring circuit, one must first understand the level of complexity of a road scene viewed from a camera.

The variability of road scenes is actually much more than what a normal person comes to imagine. Indeed, a color image, which has eight bits for each colors (then, 24-bit, as there are 3 colors) may encode 224 different color levels per pixel (more than 65,000 different possible values). HD video has more than 2 million pixels.
This means that the matrix of HD 8-bit color image may encode more than 65 0002 000 000 images !
This huge number is simply unimaginable.

Read the entire paper

Accidented Tesla
http://www.nextinpact.com/news/100804-accident-mortel-tesla-model-s-roulait-trop-vite-enquete-se-poursuit.htm

Nexyad RoadNex v2.2 – Road Detection on little road in Forest

Nexyad RoadNex v2.2 – Road Detection on little road in Forest
RoadNex detecting the lane on the road with or without markings.
The green Arrow shows the possible direction to follow.

Nexyad provides modules for ADAS (Advanced Driver Assistance Systems) : some of those modules such as RoadNex road detection or ObstaNex obstacles detection are competitors of the famous modules of the company Mobileye.

RoadNex v2.2 Scooter passing by in City Traffic

RoadNex v2.2 Road Detection – Scooter passing by in City Traffic

RoadNex v2.2 Road Detection – Scooter passing by in City Traffic
RoadNex detecting the lane on the road with or without markings.
Here, see the red line that reduces with scooters and cars; and the red surface highlighted which stops with scooters and cars, telling us it’s not road anymore, so detecting obstacles in negative.
The green Arrow shows the possible direction to follow.

Nexyad provides modules for ADAS (Advanced Driver Assistance Systems) : some of those modules such as RoadNex road detection or ObstaNex obstacles detection are competitors of the famous modules of the company Mobileye.

RoadNex v2.2 Road Detection for ADAS : Green Desert Track

NEXYAD RoadNex v2.2 Road Detection for ADAS : Green Desert Track
RoadNex detects the lane and the road surface and does not depend on the level of equipment of road infrastructure. Example here, on a dirt track in the middle of nowhere…



RoadNex v2.2 Road Detection for ADAS : Case of Red Road

RoadNex v2.2 Road Detection for ADAS : Case of Red Road
RoadNex detecting the lane and the surface of a road in red color.

Nexyad provides modules for ADAS (Advanced Driver Assistance Systems) : some of those modules such as RoadNex road detection or ObstaNex obstacles detection are competitors of the famous modules of the company Mobileye.

RoadNex v2.2 Road Detection for ADAS on Road at Night

Nexyad RoadNex v2.2 Road Detection for ADAS on Road at Night
RoadNex detecting the lane and the surface of the road at night when the light are on.

Nexyad provides modules for ADAS (Advanced Driver Assistance Systems) : some of those modules such as RoadNex road detection or ObstaNex obstacles detection are competitors of the famous modules of the company Mobileye.

RoadNex v2.2 Road with markings on both sides

RoadNex v2.2 Road with markings on both sides.
RoadNex detecting the lane on road and detecting the surface of the road with strong sun light on front of the car and the camera…

Nexyad provides modules for ADAS (Advanced Driver Assistance Systems) : some of those modules such as RoadNex road detection or ObstaNex obstacles detection are competitors of the famous modules of the company Mobileye.

RoadNex v2.2 Desert Track

RoadNex v2.2 Desert track
RoadNex detecting the lane on the desert track : sand road, no markings, stones, etc…
Contrast of colors is very poor, detection is still fine.

Nexyad provides modules for ADAS (Advanced Driver Assistance Systems) : some of those modules such as RoadNex road detection or ObstaNex obstacles detection are competitors of the famous modules of the company Mobileye.

Nexyad dans le journal Le Monde

La voiture autonome fait rêver la filière française

Un article du journal Le Monde.

«Avec les autres sociétés, nous rassemblons environ 150 personnes et réalisons tous ensemble un millions d’euros de chiffre d’affaires, soit une entreprise de taille intermédiaire. Cela nous permet de démarcher les équipementiers et les constructeurs, et de les rassurer sur notre pérennité. Seul, c’est beaucoup plus compliqué», explique Gérard Yahiaoui, le PDG de Nexyad, une PME qui développe des algorithmes capables d’agréger les images captées par les caméras afin de «voir» la route. C’est surtout l’un des principaux concurrents de la start-up israélienne Mobileye, le leader mondial de ce nouveau marché.

Nexyad Le Monde

NEXYAD in French Magazine

NEXYAD was compared to Mobileye and considered as a serious player in the competition.
Magazine : Le Journal de l’Automobile, pp 52-54, 18 Sept 2015
« MobilEye a de la concurrence : longtemps en position monopolistique, la société israélienne a désormais un
concurrent qui s’annonce sérieux dans le domaine des algorithmes de gestion des caméras embarquées, Nexyad.
Rencontre avec les ingénieurs français qui pourraient changer la donne »

http://www.journalauto.com/lja/article.view/23321/mobileye-a-de-la-concurrence/6/services